Cambridge University Press

978-1-107-02097-9 - Digital Waveform Generation

Pete Symons
Frontmatter
More information

Digital Waveform Generation

This concise overview of digital signal generation will introduce you to powerful,
flexible and practical digital waveform generation techniques. These techniques, based
largely on phase accumulation and phase–amplitude mapping, will enable you to
generate sinusoidal and arbitrary waveforms in real-time with independently controlled
waveshape, frequency, phase offset and amplitude, and to design and optimise bespoke
digital waveform generation systems from scratch.
The book includes a review of key definitions, a brief explanatory introduction to
classical analogue waveform generation and its basic conceptual and mathematical
foundations, coverage of recursion, DDS, IDFT and dynamic waveshape and spectrum
control, and a chapter dedicated to detailed examples of hardware design, accompanied
by downloadable Mathcad models created to help you explore ‘what if?’ design
scenarios. It is essential reading for practitioners in the digital signal processing
community, and for students who want to understand and apply digital waveform
synthesis techniques.

Pete Symons is a professional engineer with over 30 years’ experience in the design of
digital and analogue signal processing systems, and is Chief Engineer at Avalon
Sciences Ltd. He has held numerous positions in industry, including Chief Engineer
for the Special Projects Research Group at Thales UK, and Group Leader of the
Electronics Research Group at the Atomic Weapons Establishment, UK. He is a
Chartered Engineer, and a Fellow of the IET.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

‘Symons writes with admirable precision and attention to detail in describing techniques
for arbitrary waveform generation based upon Direct Digital Synthesis principles.
Newcomers to the topic and seasoned practitioners alike will benefit from the extensive
tutorial content supporting the theoretical foundations, synthesis algorithms and pro-
cessing architectures employed in the implementation of these powerful and flexible
techniques. A notable feature is the inclusion of simulated performance results for
several arbitrary waveform phase–amplitude mapping algorithms, illustrating the
dependence of the amplitude spectrum and other performance metrics on a range of
design and control parameters.’
Mike Meade, Open University

‘This is a very readable signal processing book – comprehensive in scope, and thorough
in presentation. Clear explanations, detailed illustrations, and practical implementation
guidance make it a welcomed addition to the literature of modern-day digital waveform
generation.’
Rick Lyons, Besser Associates

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

Digital Waveform Generation

PETE SYMONS

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

University Printing House, Cambridge CB2 8BS, United Kingdom

Published in the United States of America by Cambridge University Press, New York

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning and research at the highest international levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781107020979
© Cambridge University Press 2014
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2014
Printed in the United Kingdom by Clays, St Ives plc
A catalogue record for this publication is available from the British Library.
Library of Congress Cataloging-in-Publication data
Symons, Pete, Dr.
Digital waveform generation / Pete Symons
Includes bibliographical references and index.
ISBN 978-1-107-020979 (Hardback)
1. Signal processing–Digital techniques. 2. Electric waves. 3. Oscillators, Electric. I. Title.
TK5102.9.S96 2014
621.38150 33–dc23 2013007041
ISBN 978-1-107-020979 Hardback
Additional resources for this publication at www.cambridge.org/symons
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

Contents

Preface page xi
Acknowledgements xvi
Glossary of terms xvii

1 Introduction to waveform generation 1

1.1 Preliminaries 3
1.1.1 Outline chapter content 3
1.1.2 Digital signal processing 5
1.1.3 Periodic and aperiodic waveforms 6
1.1.4 Introducing the sine wave – properties and parameters 9
1.1.5 Instantaneous phase and frequency 10
1.1.6 From phase to amplitude – the waveform function 12
1.1.7 Signal definition – waveform or spectrum? 14
1.1.8 Signal specification – time or frequency domain? 15
1.1.9 A brief history of digital waveform generation 17
1.2 A taxonomy of electronic waveform generation 19
1.2.1 Background 19
1.2.2 Analogue waveform generation 19
1.2.3 Hybrid waveform generation 23
1.2.4 Digital waveform generation 24
1.2.5 Arbitrary waveform generation 25
1.3 Analogue waveform generation 31
1.3.1 Analogue waveform generation in test instrumentation 31
1.3.2 The Wien bridge oscillator 32
1.3.3 The state variable oscillator 34
1.3.4 The voltage-controlled switched integration oscillator 36
1.3.5 The charge-balancing voltage to frequency convertor 38
1.3.6 The shortcomings of analogue waveform generation 38
1.4 Applications of arbitrary waveform generation 41
1.4.1 Test and diagnostic applications 41
1.4.2 Signal emulation applications 42
1.4.3 Bespoke embedded applications 43
1.4.4 RF up-conversion applications 43

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

vi Contents

1.4.5 Unique DDS AWG attributes 45

1.4.6 A simple example – testing a PLL 45
1.5 Summary 47
1.6 References 48

2 The foundations of digital waveform generation 49

2.1 Mathematical preliminaries 49
2.1.1 Continuous- and discrete-time signals 49
2.1.2 The phase–frequency relationship of a discrete-time sinusoid 52
2.1.3 Analogue reconstruction and alias image rejection 54
2.1.4 Quantisation and digital to analogue conversion 54
2.2 A fundamental concept – sampling a tabulated signal 55
2.2.1 Tabulating a signal – introducing the wavetable 55
2.2.2 Regenerating a tabulated signal 56
2.2.3 A sample rate conversion view 57
2.2.4 Frequency control resolution and wavetable length 58
2.3 Tabulating a waveform function 59
2.3.1 The wavetable revisited 59
2.3.2 Tabulating the sine function 60
2.3.3 Time domain arbitrary waveform specification 62
2.3.4 Frequency domain arbitrary waveform specification 63
2.4 The foundations of direct digital synthesis (DDS) 64
2.4.1 Phase accumulation frequency synthesis 64
2.4.2 Phase–amplitude mapping 67
2.4.3 Phase continuity 68
2.4.4 Phase–amplitude mapping and fractional addressing 70
2.4.5 The aliasing problem 73
2.5 Control parameters and their characteristics 74
2.5.1 Desirable control characteristics 75
2.5.2 Frequency control 76
2.5.3 Phase offset control 78
2.5.4 Amplitude control 79
2.5.5 Waveshape control 79
2.5.6 Spectrum control 81
2.6 Performance metrics 83
2.6.1 Phase truncation noise 83
2.6.2 The amplitude error spectrum 85
2.6.3 Waveshape error 85
2.6.4 Phase noise 86
2.6.5 Arithmetic processing requirements 88
2.7 Summary 89
2.8 References 89

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

Contents vii

3 Recursive sine wave oscillators 90

3.1 Mathematical preliminaries 91
3.2 The direct-form 93
3.3 The coupled-form 99
3.4 The modified coupled-form 101
3.5 The waveguide-form 101
3.6 Implementation considerations 102
3.6.1 Initial conditions 102
3.6.2 Arithmetic quantisation effects 104
3.6.3 Frequency control 106
3.7 Summary 107
3.8 References 108

4 DDS sine wave generation 109

4.1 Phase accumulation, truncation and arithmetic processing 109
4.1.1 Ideal sinusoidal DDS 110
4.1.2 Optimal phase–amplitude mapping 112
4.1.3 Phase truncation and fractional wavetable addressing 114
4.1.4 Linear phase interpolation 122
4.1.5 Static and dynamic phase offset 124
4.1.6 Quadrature sine wave generation 125
4.1.7 Introducing phase domain processing 127
4.2 Theoretical analysis of DDS phase truncation errors 128
4.2.1 The DDS error model 128
4.2.2 Phase truncation errors 129
4.2.3 Wavetable sample quantisation errors 132
4.3 Sinusoidal phase–amplitude mapping 134
4.3.1 SNR and oversampling 134
4.3.2 Phase truncated wavetable indexing 135
4.3.3 Phase dithered wavetable indexing and noise shaping 138
4.3.4 Linear phase interpolated wavetable indexing 144
4.3.5 Trigonometric identity phase interpolation 148
4.3.6 Sinusoidal wavetable compression 152
4.3.7 Direct computation phase–amplitude mapping – the CORDIC
algorithm 155
4.4 Summary 159
4.5 References 161

5 DDS arbitrary waveform generation 162

5.1 From sinusoidal to arbitrary waveforms 162
5.1.1 Phase truncation and the need for interpolation 163
5.1.2 Introducing Lagrange polynomial interpolation 165

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

viii Contents

5.1.3 The harmonic aliasing problem 170

5.1.4 Time domain specification of an arbitrary waveform 172
5.1.5 Frequency domain specification of an arbitrary waveform 178
5.1.6 Defining DDS AWG simulation parameters 181
5.1.7 Optimal phase–amplitude mapping 184
5.2 Interpolated phase–amplitude mapping 186
5.2.1 Phase truncated wavetable indexing 186
5.2.2 Phase dithered wavetable indexing 187
5.2.3 Linear phase interpolated wavetable indexing 192
5.2.4 Lagrange polynomial phase interpolated wavetable indexing 195
5.3 Analogue waveshaping 203
5.3.1 The phase accumulator as a digital frequency synthesiser 205
5.3.2 Analogue waveshaping principles and some fundamental
problems 207
5.3.3 Triangle waveform shaping 211
5.4 DDS clock generation 213
5.4.1 Hard-limited sinusoidal DDS 214
5.4.2 Reconstruction filter bandwidth 216
5.5 Variable sample frequency AWG 217
5.5.1 Basic architecture and implementation limitations 217
5.5.2 Harmonic aliasing and reconstruction filter considerations 218
5.6 Computer music and audio test applications of DDS AWG 219
5.6.1 Harmonics and partials 220
5.6.2 Frequency control, musical pitch and tuning 221
5.6.3 Audio measurement – harmonic and intermodulation distortion 222
5.7 Summary 225
5.8 References 227

6 Dynamic waveshape and spectrum control 228

6.1 Paged wavetable memory – piecewise stepped waveshaping 229
6.1.1 Page indexing a group of wavetables 229
6.1.2 Specifying a group of wavetables 231
6.1.3 Paged wavetable interpolation – piecewise continuous
waveshaping 233
6.1.4 Frequency domain behaviour 235
6.2 Linear combination of multiple wavetables 236
6.2.1 Single phase accumulator case – harmonic signal synthesis 237
6.2.2 The IDFT as a limiting case 239
6.2.3 Multiple phase accumulator case – band-pass signal synthesis 241
6.3 Modulation 242
6.3.1 Frequency, phase and amplitude modulation 243
6.3.2 FSK and PSK modulation 244
6.3.3 Frequency sweep and chirp generation 246

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

Contents ix

6.4 Summary 247

6.5 References 248

7 Phase domain processing – DDS and the IDFT 249

7.1 Fundamental concepts 249
7.1.1 The inverse discrete Fourier transform (IDFT) 250
7.1.2 Weighted phase sequence summation 252
7.1.3 Dynamic spectrum control and the metaparameter 254
7.1.4 Aliasing control and the mitigation of Gibbs phenomenon 256
7.1.5 Time division multiplexing (TDM) 257
7.2 Contiguous harmonic phase generation and the IDFT 258
7.2.1 Harmonic phase accumulation 258
7.2.2 Harmonic phase offset 260
7.2.3 Amplitude accumulation 261
7.3 Non-contiguous harmonic phase generation 262
7.3.1 Phase multiplication 263
7.3.2 Modulo integer multiplication 263
7.4 Summary 264
7.5 References 265

8 Hardware implementation architectures 266

8.1 Processing techniques and design considerations 267
8.1.1 Pipelined and parallel processing 267
8.1.2 Sample and block pipelining 268
8.1.3 Time division multiplexing 272
8.1.4 Arithmetic and modulo accumulation 273
8.1.5 High speed binary addition 274
8.1.6 Wavetable memory architectures 275
8.1.7 Vector memory 277
8.2 Partitioned phase accumulation 280
8.3 Sinusoidal phase–amplitude mapping 283
8.3.1 Linear phase interpolation 284
8.3.2 Trigonometric identity phase interpolation 286
8.4 Interpolated vector memory 289
8.4.1 Linear interpolation architecture 289
8.4.2 Cubic interpolation architecture 289
8.4.3 Interpolated vector memory design considerations 293
8.4.4 A paged wavetable interpolation memory 296
8.5 Waveform generation by direct computation 296
8.5.1 A contiguous harmonic IDFT waveform synthesiser 299
8.5.2 An arbitrary harmonic IDFT waveform synthesiser 302
8.5.3 A band-pass spectrum waveform synthesiser 302

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

x Contents

8.6 Summary 304

8.7 References 306

9 Digital to analogue conversion 307

9.1 Error mechanisms 311
9.1.1 Offset error 313
9.1.2 Gain error 315
9.1.3 Settling time 315
9.1.4 Transition or ‘glitch’ errors 317
9.1.5 Linearity errors 317
9.1.6 DAC quantisation noise 318
9.2 DAC dynamic specification considerations 319
9.2.1 Harmonic and intermodulation distortion 319
9.2.2 Spurious-free dynamic range (SFDR) 321
9.2.3 Signal-to-noise ratio (SNR) 321
9.3 The reconstruction filter 322
9.3.1 Frequency and phase response 323
9.3.2 Filter implementation – passive or active? 325
9.3.3 Signal distortion pitfalls – maintaining SFDR 327
9.3.4 Reconstruction filter design considerations – hints and tips 328
9.4 Supplementary analogue design considerations 330
9.4.1 The DAC reference voltage and output amplifier 330
9.4.2 Switched attenuation and output level control 332
9.4.3 DC offset and output dynamic range considerations 335
9.4.4 The output drive amplifier 336
9.4.5 Square and pulse waveforms 337
9.5 Summary 338
9.6 References 340

Index 341

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

Preface

Most electronic design engineers, irrespective of being the ‘analogue’ or ‘digital’

variety, are occasionally faced with the task of designing an oscillatory signal generator
with particular implementation constraints, control and performance requirements.
These requirements might include extremely low distortion, unusual ‘application spe-
cific’ waveshape, wide frequency tuning range, low temperature drift, and so on.
Historically, such a task will have been tackled with a wholly analogue design, possibly
augmented by digital control, where extremely high levels of performance are evident in
some cases. If we take high-end audio test instrumentation as an example, the now
legendary Hewlett Packard HP8903B and Audio Precision AP1 audio test sets both use
digitally controlled analogue state variable oscillators to generate extremely low distor-
tion sine waves. The state variable analogue oscillator is effectively an analogue
computer model designed to compute solutions of a second-order differential equation.
A specific class of solution (under certain parametric conditions) is a continuous
sinusoidal oscillation. These generators are outstanding examples of what can be
achieved with innovative analogue design. However, the world is becoming increas-
ingly digital and very high levels of digital processing power can be implemented at
relatively low cost. Various ‘all digital’ waveform generation techniques are therefore
now practicable; and when all of their advantages are weighed up against the disadvan-
tages (yes, digital processing is not necessarily a panacea to guarantee ideal perform-
ance), they nearly always represent the best solution. This approach is reinforced, if not
driven, by the ever-improving performance of commercial digital to analogue conver-
sion (DAC) devices as measured by their spurious-free dynamic range (i.e. distortion)
and bandwidth. It is not unreasonable to state that the integrated DAC is the foremost
enabling technology for nearly all applications of digital waveform generation. Excep-
tions to this observation apply to purely digital signals, which exist as a discrete-time
sequence of binary numbers representing the signal waveform samples.
Electronic signals are often described as waveforms, indicating the time domain
image observed when the signal is measured with an oscilloscope. A signal produced
by an electronic signal generator may be defined as a periodic (or aperiodic), time-
varying voltage or current. The corresponding shape function can be described (or
approximated to some level of accuracy) by a causal mathematical function. This
definition suggests a generation method where the underlying mathematical function
which describes the waveform is effectively computed in real time by analogue or
digital means. The simplest and most widely used waveform is the sine wave, whose

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

xii Preface

corresponding mathematical function may be computed by various well-known ana-

logue methods, digital recursive algorithms and so-called ‘direct digital synthesis’
(DDS). Classical non-sinusoidal, so-called ‘function’ waveforms include the sawtooth,
triangle, square and pulse waveforms; each having a simple piecewise-linear discon-
tinuous mathematical description. These waveforms can also be generated using an
analogue computer model so as to have a waveshape which changes according to a
waveshape modulation parameter. For example, a pulse waveform whose duty-cycle
(i.e. the ratio of the waveform’s ‘on’ time to the total waveform period) changes
according to the value of some parameter, although the waveform’s frequency remains
constant. These waveforms suffice for many applications, but increasingly signals with
user-defined waveshape (or spectrum) are required in complex electronic systems as test
stimuli, excitation, control or modulation signals. We generalise signals with user-
defined waveshape as arbitrary waveforms and this book is primarily concerned with
investigating digital techniques for their generation. Further examples of arbitrary
waveform application include:

 communication system carrier modulation;

 transducer excitation signal generation;
 medical and related imaging signal emulation;
 signal emulation in complex test instrumentation;
 radar return signal simulation;
 electronic warfare adaptive signal generation;
 control system diagnostic excitation;
 multi-tone audio system testing;
 computer music synthesis.
Many books have been written on signal generation by purely analogue means, such as
phase shift and state variable oscillators or analogue computer function generators.
Indeed, the first electronic signal generators, whether laboratory instrumentation or
bespoke, were completely analogue systems, and only much later augmented by various
forms of digital control. Over time, these techniques evolved to allow the generation
of so-called ‘function waveforms’ (e.g. triangle, sawtooth and pulse) in addition to
the ubiquitous sine wave. Indeed, some novel designs were capable of generating
rudimentary arbitrary waveforms by purely analogue means according to a piecewise-
linear shape specification. The HP3314A is one example of such a commercial labora-
tory signal generator. However, analogue techniques have always been beset by the
same fundamental problems – thermal instability, the need for periodic recalibration,
lack of control flexibility and limited waveform shape or corresponding spectrum
specification.
Before the emergence of DSP techniques, non-sinusoidal waveforms were generated
by analogue computation models such as the voltage controlled function generator.
Subsequently, digital control of key parameters was realised by using a digital to
analogue convertor (DAC) to translate from the digital to analogue domain, allowing
computer control of the signal generation process. Today, digital signal generation is
a cornerstone function in many modern analogue and DSP systems. Novel DSP

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

Preface xiii

techniques enable the generation of essentially any waveform shape, and provide
independent control of waveshape, frequency, phase offset and amplitude. These
techniques have evolved from the well-established methods of phase accumulation
and phase–amplitude mapping, which have hitherto been widely applied to the gener-
ation of sine wave signals through DDS. Recursive digital techniques have also
been applied to sine wave synthesis, but while having distinct advantages in some
applications, they lack the flexibility of the phase accumulation approach. They
are, however, well suited to many application areas due to their simplicity and low
computational cost.
This book investigates signal generation techniques based on phase accumulation and
phase–amplitude mapping. These ‘all-digital’ techniques enable the generation of
sinusoidal and arbitrary waveforms with user-programmable waveshape, frequency,
phase offset and amplitude. These powerful, yet flexible, techniques synthesise wave-
forms whose shape (and hence corresponding spectrum) can be defined according to a
time or frequency domain specification. Furthermore, with appropriate phase–amplitude
mapping, the synthesised waveshape can be varied according to a control parameter in
real time. Multiple waveform generators are easily synchronised, and with program-
mable phase offsets generate polyphase waveforms or quadrature IQ sine wave constel-
lations. The true utility of arbitrary waveform generation can be appreciated when we
consider that a waveform’s instantaneous amplitude may correspond to any parameter
in a digital or analogue system, whether as a test or control signal. For example, to
control the set point of a servo mechanism, the instantaneous amplitude or phase of a
carrier signal or the forcing function is applied to evaluate a control system’s response.
Motivation for this book has come in part from observations of the currently available
signal generation literature. Many excellent works have been written on the digital
generation of sinusoidal signals using DDS and related techniques. However, there is a
notable deficiency in the area of non-sinusoidal or arbitrary waveform generation using
purely digital techniques. Several published texts have dealt with waveform generation
in the field of computer music, and these can be considered as special cases of the
techniques described here. This book endeavours to address the gap by providing a
treatise on both sinusoidal and arbitrary waveform generation that incorporates tutorial
and hardware implementation descriptions at the functional block level. It is this latter
material which is intended to assist the engineer or system architect who has the task of
designing a bespoke waveform generator for a specific application. Synthesis algo-
rithms are presented as signal-flow block descriptions and are deliberately abstracted
from specific hardware or software implementation technologies (e.g. FPGA or DSP
code). However, generic throughput enhancement techniques such as arithmetic pipe-
lining and parallel memory architectures are discussed. Accordingly, it is assumed that
practitioners who are suitably ‘skilled in the art’ can implement and optimise these
signal flow descriptions in a hardware or software technology appropriate to their
application. It is reasonable to suggest that many of the suggested hardware architec-
tures, and the underlying synthesis techniques which they implement, are ‘solutions
looking for a problem’. Hopefully, this book will enable the reader to find these
techniques a good home, or at the very least, provide food for thought.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

xiv Preface

We begin with a review of key definitions, historical context, classical analogue

waveform generation and the conceptual and mathematical foundations pertinent to
digitally generating electronic signals. The text then presents a detailed review of sine
wave synthesis algorithms beginning with recursion and finally progressing to sinus-
oidal direct digital synthesis (DDS). This lays the essential foundations required for an
understanding of phase accumulating arbitrary waveform generation – the focus of this
book. The figure below illustrates the flow and interconnection between chapters. To
supplement the written material and to assist the reader in ‘what happens if’ design
simulation, Mathcad models are available that compute qualitative performance metrics
for various configurations of the techniques presented. Mathcad models are also used to
generate the graphical plot figures used throughout the book to assist communication of
key ideas. These models are available for free download from www.cambridge.org/
symons, and from the author’s website www.petesymons.com/dwg.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

Preface xv

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

Acknowledgements

I would wish to send my heartfelt thanks to Karen, my long-suffering wife, without

whose love and encouragement this book would certainly not have been written; to
Chris, Ben and Jenny for making me so proud; to my oldest friend Mike McNabb for
providing useful feedback on writing style and always making me laugh; and finally,
Hannah Eustace for her detailed critiques of several draft chapters from a non-technical
perspective.
I dedicate this book to the memory of my father, Bob Symons, who both started and
fuelled my love of electronics as a young boy who was always taking the back off the
television set to see how it worked; and Bob Denson – my ‘Uncle Bob’ – simply the
nicest and most decent person I have ever known.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

Glossary of terms

(Terms are grouped as used in the analysis.)

A waveform amplitude scaling factor.

B post-DAC low-pass reconstruction filter 3 dB bandwidth.
b number of amplitude sample bits.
C waveform crest factor – ratio of waveform peak to RMS amplitude.
D waveform duty-cycle – ratio of waveform ‘on’ time to period.
εa(n) amplitude error sequence.
t continuous-time variable.
f waveform or signal frequency in hertz.
τ waveform period equivalent to the reciprocal of waveform frequency in
hertz.
f0 fundamental frequency, typically of a harmonic series.
fo waveform generator output frequency.
f(t) continuous-time instantaneous frequency in hertz.
fs sampling frequency (or rate) in hertz.
T sample period (or interval) equivalent to the reciprocal of sampling
frequency in hertz.
βR the oversampling ratio between the processing sample frequency fs and an
equivalent sample frequency corresponding to twice the bandwidth of the
post-DAC reconstruction filter B.
y(t) or x(t) continuous-time waveform amplitude signal.
X( f ) continuous complex amplitude spectrum.
θ waveform phase offset (typically in radians).
ϕ(t) continuous-time instantaneous phase.
ϕw(t) wrapped continuous-time instantaneous phase.
〈a〉b ‘a modulo b‘ – the remainder after repeated subtraction of b from a until
(ab) < b.
(a, b) greatest common divisor between a and b.
M phase accumulator word length – 2M possible phase states.
I phase accumulator integer component word length.
F phase accumulator fraction component word length.
ϕ(n) instantaneous discrete-time phase sequence.

© in this web service Cambridge University Press www.cambridge.org

Cambridge University Press
978-1-107-02097-9 - Digital Waveform Generation
Pete Symons
Frontmatter
More information

xviii Glossary of terms

ϕI(n) integer component of discrete-time instantaneous phase within a fractional

phase representation.
ϕF (n) fraction component of discrete-time instantaneous phase within a fractional
phase representation.
α(n) rational fraction interpretation of ϕF(n).
φ phase increment.
φ(n) discrete-time phase increment sequence.
Λ number of sinusoidal DDS amplitude spectrum spurs.
ϑ(n) discrete-time phase dither sequence.
Ak kth element of the IDFT harmonic amplitude vector.
θk kth element of the IDFT harmonic phase offset vector.
Hk kth element of an arbitrary waveform band-limiting amplitude response.
Nh number of harmonics in a harmonic waveform specification.
Ns number of samples in a particular computation.
S[a] ath element of a sinusoidal wavetable.
W[a] ath element of an arbitrary waveform wavetable.
L wavetable length in samples.
PN (x) Nth-order Lagrange interpolation polynomial.
P number of bits in a wavetable page address – 2P wavetables.
p(n) wavetable page address sequence.
PP (n) P-bit integer wavetable page index sequence component.
γ(n) F 0 -bit fraction wavetable page index sequence component.

© in this web service Cambridge University Press www.cambridge.org